The present invention generally relates to image processing. More specifically, the present invention relates to systems and methods for image processing of two-dimensional (2D) medical images.
Hospitals and other medical facilities, such as clinics and imaging centers, continually seek to improve or optimize utilization of resources and productivity. Such optimization may be achieved though faster or more understandable image processing. For example, improved image processing may include faster calculations, lower usage of computer hardware graphics, and lower storage requirements. Further, simplified images, better optimization of a single screen, and undistorted images may each improve productivity and can lead to decreased mistakes.
Many techniques are used currently to process images involving image slices. Image slices are 2D images that may be created using a tomographic scan of an object, for example. In a tomographic scan, image slices may be taken at intervals along the object. For example, each image slice may represent a single 2D axial cross-section of the object.
Image slices are not easy to interpret. From a 2D image, it is difficult to see where the image is located in the object. One common solution is to draw a cut line on a localizer image. A localizer image may be a rendition of the object that the image slices are created from. Cut lines, also known as scout-lines, are lines drawn across a localizer image. The cut line indicates where the image slice is drawn from. Typically, the localizer image and the image slice are displayed side-by-side, with the cut line on the localizer image indicating the plane in which the image slice was taken.
Although cut lines are useful, this method forces the user to divide their attention between two views. Further, there is a risk of mistakes. For example, the viewer may confuse the left-right orientation of the object. As another example, cut lines are not easy to understand by casual users, such as referring physicians and patients.
Another method of viewing image slices is full 3D image volume rendering. This method cuts down on the risk of some mistakes because the 3D visualization may be more intuitive to interpret certain aspects such as left/right. However, 3D image rendering requires 3D reconstruction which involves increased computational power and is slow to calculate. This method also may be confusing to a user due to the amount of information that is displayed. That is, a user may experience information overload. In addition, abnormalities in images are often more clearly visible in the 2D slices, and the 2D view is therefore the preferred view.